BIMIM: Band-Independent Masked Image Modeling With Transformer for Multispectral Satellite Imagery
文献类型：期刊论文
| 作者 | Song, Jia2,3; Xia, Luosheng1,2 |
| 刊名 | IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
 |
| 出版日期 | 2026 |
| 卷号 | 19页码:6443-6459 |
| 关键词 | Transformers Image reconstruction Remote sensing Feature extraction Satellite images Computational modeling Adaptation models Representation learning Land surface Training Embedding masked image modeling (MIM) multispectral image classification self-supervised learning (SSL) transformer |
| ISSN号 | 1939-1404 |
| DOI | 10.1109/JSTARS.2026.3660330 |
| 产权排序 | 1 |
| 文献子类 | Article |
| 英文摘要 | Self-supervised learning (SSL) offers a promising solution to reduce reliance on labeled data. Among SSL approaches, Masked Image Modeling (MIM) has demonstrated significant potential in remote sensing applications such as scene classification and semantic segmentation, owing to its ability to capture pixel-level details. However, existing MIM frameworks, originally designed for natural images, struggle to adapt to the spectral-spatial characteristics of multispectral satellite imagery. While recent studies have introduced spectral-enhanced MIM SSL methods, most rely on band-group embedding, which imposes constraints on band utilization flexibility in downstream fine-tuning tasks and limits the granularity of spectral feature learning. To address these challenges, this study proposes Band-Independent Masked Image Modeling (BIMIM) with Transformer, a novel SSL framework specifically designed for multispectral satellite imagery. BIMIM not only enables finer band-specific spectral feature extraction, allowing for more effective capture of subtle spectral variations, but also introduces spatially random masking at the single-band level, facilitating more efficient interband feature learning. Extensive experiments on publicly available remote sensing datasets demonstrate that BIMIM achieves state-of-the-art performance in downstream tasks such as scene classification and semantic segmentation. This study provides a new perspective on SSL for multispectral remote sensing, paving the way for more effective spectral-spatial feature extraction and adaptation in SSL frameworks. |
| URL标识 | 查看原文 |
| WOS关键词 | LAND-COVER ; WATER |
| WOS研究方向 | Engineering ; Physical Geography ; Remote Sensing ; Imaging Science & Photographic Technology |
| 语种 | 英语 |
| WOS记录号 | WOS:001696554500005 |
| 出版者 | IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC |
| 源URL | [http://ir.igsnrr.ac.cn/handle/311030/221361]  |
| 专题 | 资源与环境信息系统国家重点实验室_外文论文 |
| 通讯作者 | Song, Jia |
| 作者单位 | 1.Univ Chinese Acad Sci, Beijing 100049, Peoples R China 2.Chinese Acad Sci, Inst Geog Sci & Nat Resources Res, State Key Lab Resources & Environm Informat Syst, Beijing 100101, Peoples R China; 3.Jiangsu Ctr Collaborat Innovat Geog Informat Resou, Nanjing 210023, Peoples R China; |
| 推荐引用方式 GB/T 7714 | Song, Jia,Xia, Luosheng. BIMIM: Band-Independent Masked Image Modeling With Transformer for Multispectral Satellite Imagery[J]. IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING,2026,19:6443-6459. |
| APA | Song, Jia,&Xia, Luosheng.(2026).BIMIM: Band-Independent Masked Image Modeling With Transformer for Multispectral Satellite Imagery.IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING,19,6443-6459. |
| MLA | Song, Jia,et al."BIMIM: Band-Independent Masked Image Modeling With Transformer for Multispectral Satellite Imagery".IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING 19(2026):6443-6459. |
入库方式： OAI收割
来源：地理科学与资源研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。